================================================================================ IBIS INTERCONNECT TASK GROUP Mailing list: ibis-interconnect@freelists.org ================================================================================ Attendees from January 24, 2024 Meeting (* means attended at least using audio) ANSYS Curtis Clark Broadcom James Church Intel Corp. Michael Mirmak* Michael Brownell Keysight Technologies Ming Yan Marvell Steve Parker MathWorks Walter Katz* Micron Technology Justin Butterfield Siemens EDA Weston Beal*, Arpad Muranyi*, Randy Wolff* ST Microelectronics Aurora Sanna Synopsys Ted Mido, Edna Moreno Teraspeed Labs Bob Ross* University of Illinois Jose Schutt-Aine Zuken USA Lance Wang* Michael Mirmak convened the meeting and took minutes. No patents were declared. During review of the minutes of the January 17 meeting, Randy Wolff moved to approve; Lance Wang seconded. The minutes were approved without objection. During the AR review, Michael noted that his AR is still open regarding checking the sampling approach used in IEEE 370. No other ARs were open. Arpad Muranyi noted that draft 18 of the pole-residue TSIRD proposal was shown last week and sent out; draft 19 was prepared, but not distributed, to make a single word change. The team focused on the Begin/End Data Source Conditions section. Arpad has changed "min frequency" to "min source frequency". He noted that the team may also want to add "min fitted frequency" though that is not now present in the draft. Michael asked about which parameters were required vs. optional. Arpad noted that Begin Data Source Conditions is required if [Begin Pole-Residue Data] is present. Bob Ross suggested making it optional. Arpad replied that the list of subparameters is "up for grabs" for changes - it's just a placeholder. Michael made some observations about spacing of the points, noting we have similar concerns when discussing "most changing" areas in IBIS tables. Arpad noted that Walter proposed linear and logarithmic in addition to adaptive as spacing types for the draft. Walter refined the suggested that the main categories are linear and logarithmic; the rest are "non-uniform" and may be called such. Arpad noted that the keyword text now includes a list of source frequencies. Weston Beal noted that this may be hundreds of frequencies long. Bob stated that he doesn't like the list of source frequencies; further, we don't need logarithmic spacing to be called out, either. Walter asked about quality metrics, noting that these are defined in IEEE 370, including passivity, causality, etc. Michael added that broadband SPICE fitting algorithms have some, but would be a new area to define in the specification. Walter suggested the specification may need to include them. Bob noted that the recipient may not be able to do much without the original data. Also, he reminded the team that Dr. Yuri Shlepnev presented on quality some time ago, specifically related to S-parameters. There is a patented quality metric owned by Keysight; tread lightly here. Pole-residue is a way of interpolating between data anyway. Arpad responded that pole-residue is actually a continuous algebraic function. Walter suggested keeping the maximum valid frequency value but removing other parameters. Bob recommended getting rid of the source frequency list. Walter moved to add a keyword or subparameter "Max valid frequency". This is not the same as max fitted frequency. Randy Wolff asked whether this would be automatically added. Michael suggested that this was a maximum recommended frequency, and that some human will have to fill this in. Bob added that someone would end up making a judgment about an equation that continues to infinity. Walter advocated getting rid of everything beyond max valid frequency. Arpad asked whether this applied to the minimum frequency (DC) too. Bob and Walter noted that zero frequency was not required in the current draft. Arpad asked about combining Data Source and Data Conditions keywords. Randy and Bob expressed opinions favoring this approach. Bob added that he would like to keep min and max valid frequency optional. Walter noted that he will circulate the document when the details are more solid, to obtain feedback. Arpad asked about upper half/lower half matrix specification compared to indexing to reuse data for compression. Randy asked about revising the existing Touchstone keyword as opposed to writing something new. Bob advocated reusing the concept; there's no error checking if there is any problem with the indexing. Randy stated that, if the intention is to have an upper-half matrix, but you don't have a keyword and only use indexing, you can make a mistake easily. We can still include indexing, but the upper half matrix approach is very handy. The team continued discussing matrix indexing vs. formatting in the draft. Arpad noted that there are as many pole-residue keywords as there are elements in the matrix. Bob suggested one could get nearly a one-half size reduction using upper/lower half matrix concepts. Randy replied that Touchstone would need some minor changes to the Matrix Format keyword to spell out the rules and expansion beyond Network Data applications. It could be easier to write up a new keyword Walter moved to avoid a new keyword and just use the matrix identifier {(1,2) etc.}. Randy noted that the matrix will always be generated by some tool; we can rely on proper terms being written out. Arpad suggested that formatting is a concern for larger matrices. Randy noted that even with this, the change still represents a huge reduction in size. Bob suggested supporting upper and lower only. Randy added that upper/lower support requires error checking; Walter suggested that we can support both. Arpad stated that the team would need to analyze how upper/lower matrix data would interact with the indexing mechanism. Walter suggested putting in index restrictions if upper/lower was used. Arpad asked about making use of the existing Touchstone keyword. Next time, the team will review draft 19. Arpad moved to adjourn; Bob seconded. The meeting adjourned without objection. The next meeting will be held on February 7, 2024. ================================================================================ Bin List: 1) [Complete draft Touchstone document separating version 1.0 and 2.0 into their own chapters] - REMOVED 2) Create structures to encapsulate Touchstone 1.0 data in Touchstone 2+ specifications - TABLED 3) Complete draft Touchstone 2.0 document containing TSIRD3 and TSIRD4 draft (Muranyi) – COMPLETED IN DRAFT 10 4) Complete pole-residue format BIRD draft (Muranyi) 5) Complete port naming proposal (Katz) 6) Create alternatives to the Touchstone 1.0 option line before the "R" character - TABLED 7) Complete ISS-IRD 1 Draft - Enable Cascading of S-parameters Through W-element (Mirmak) - TABLED 8) Complete/revise Touchstone 3.0 draft outline (Mirmak) – dependent on several items above Tabled ARs: - Arpad to give an example of the physical connectivity needed for EMD automation.